Wikimedia Research - Programs - Knowledge Gaps

A multilingual model for entity insertion in Wikipedia articles Nov 2024 We published a new paper at EMNLP ‘24 where develop a model for automatically locating a suitable position for a new link in a Wikipedia article which could support editors in cases where a suitable anchor text does not yet exist. arxiv.org > abs > 2410.04254 Curiosity of Wikipedia readers Oct 2024 We published a new paper in Science Advances where we uncover complex patterns of Wikipedia navigation and characterizes reader curiosity types. science.org > doi > 10.1126 > sciadv.adn3268 A multilingual model for measuring readability Aug 2024 We published a new paper at ACL ‘24 where we develop a multilingual model to score the readability of Wikipedia articles across languages. arxiv.org > abs > 2406.01835 Language-Agnostic Modeling of Wikipedia Article Quality Apr 2024 We published a dataset paper in ICWSM '24 with the results of applying our language-agnostic article quality model to millions of revisions from over 300 language editions of Wikipedia. arxiv.org > abs > 2404.09764 Recommender Systems to Reduce Content Gaps Apr 2024 We published a paper in ICWSM '24 describing an experiment to explore the potential role of recommender systems in reducing content gaps on Wikipedia. arxiv.org > abs > 2307.08669 Impact of the Newcomer Homepage Aug 2023 We published a new paper showing how the Newcomer Homepage has increased participation amongst newcomers to Wikipedia. arxiv.org > abs > 2308.09642 Temporal Regularities of Wikipedia Consumption Jul 2023 We published a new paper about the temporal patterns in how Wikipedia articles are accessed by readers helping us understand the diversity of their information needs. arxiv.org > abs > 2305.09497 Orphan Articles: The Dark Matter of Wikipedia Jul 2023 We published a new paper about the surprisingly large number of orphan articles in Wikipedia and how to improve their visibility. arxiv.org > abs > 2306.03940 Leveraging Recommender Systems to Reduce Content Gaps on Peer Production Platforms Jul 2023 We published a new paper about editor willingness to accept more diverse recommendations that shows promise for using recommender systems to address content gaps. arxiv.org > abs > 2307.08669 Architectural styles of curiosity in global Wikipedia mobile app readership doi.org > 10.1126 > sciadv.adn3268 Entity Insertion in Multilingual Linked Corpora: The Case of Wikipedia arxiv.org > abs > 2410.04254 Language-Agnostic Modeling of Wikipedia Articles for Content Quality Assessment across Languages arxiv.org > abs > 2404.09764 Leveraging Recommender Systems to Reduce Content Gaps on Peer Production Platforms arxiv.org > abs > 2307.08669 Orphan Articles: The Dark Matter of Wikipedia arxiv.org > abs > 2306.03940 Curious Rhythms: Temporal Regularities of Wikipedia Consumption arxiv.org > abs > 2305.09497 Increasing Participation in Peer Production Communities with the Newcomer Homepage arxiv.org > abs > 2308.09642 https://doi.org/10.1145/3610071 doi.org > 10.1145 > 3610071 A Large-Scale Characterization of How Readers Browse Wikipedia arxiv.org > abs > 2112.11848